Dynamic contrast-enhanced MRI (DCE-MRI) plays a critical role in the diagnosis of tumors and in assessing their response to therapy. High temporal resolution must be achieved to accurately determine tumor enhancement kinetics. In addition, sufficient spatial resolution is needed for the identification and tracking of small lesions and for assessing intratumoral heterogeneity. Furthermore, there is a potential need for large volume or whole-organ imaging for the evaluation of multiple metastatic lesions. Unfortunately, using current techniques, there are tradeoffs between high temporal resolution, high spatial resolution, and volumetric coverage. It is therefore difficult to acquire a dynamic MR imaging series able to simultaneously meet these diverging criteria in a single scan. Moreover, respiratory motion further complicates the problems when applying DCE-MRI in patients with metastatic disease in the chest or the abdomen. With existing technology, it is therefore extremely difficult to accurately assess perfusion of lung or liver tumors. Here we propose to develop, implement, and validate novel strategies for improved quantification of tumor contrast kinetics in 3D DCE-MRI, with the goal of achieving in a single dynamic series, high temporal and spatial resolutions, in conjunction with effective motion compensation. The proposed work is based on radial data acquisition and reconstruction strategies based on the golden-angle view order scheme, which permit (1) highly flexible reconstruction schemes allowing for enhanced temporal resolution;and (2) effective self-gating strategies for respiratory motion compensation. Our overall hypothesis is that the proposed radial DCE-MRI methodology will significantly improve the reliability of tumor perfusion measurements in regions where respiratory motion can seriously hamper accurate analysis. The specific aims of this proposal are as follows: Specific Aim 1: Explore novel self-gating strategies for compensation of respiratory motion in radial DCE- MRI. We will investigate improved methods to enhance the temporal resolution of respiratory self-gated radial DCE-MRI and investigate techniques to compensate for residual motion. Specific Aim 2: Develop optimized image reconstruction strategies for accurate tumor perfusion assessment. We will develop strategies for image reconstruction that optimizes perfusion measurement accuracy. We will also examine methods to more fully automate the image reconstruction process. Specific Aim 3: Evaluate the proposed radial versus conventional rectilinear acquisition schemes for DCE-MRI evaluation in subjects with benign liver lesions. A randomized study will be carried out to test the hypothesis that the proposed methodologies outlined in Aims 1 and 2-when compared against conventional imaging methods for DCE-MRI image acquisition-will yield data sets with superior image quality, and will result in more reproducible DCE-MRI quantification of tumor perfusion. PUBLIC HEALTH RELEVANCE: Dynamic contrast-enhanced MRI plays a critical role in the diagnosis of tumors and in assessing their response to new therapies. However, due to the tradeoffs among high spatial resolution, high temporal resolution, and volumetric coverage, it is difficult with existing techniques to assess lesions accurately, particularly those that are located in the chest or abdomen where respiratory motion further complicates measurement. The proposed approach provides a means for accurate assessment of tumor perfusion by permitting high spatial and temporal resolution imaging with large volume coverage, in conjunction with an effective strategy for respiratory motion compensation.